DESCRIPTION: (Applicant's Abstract) Immunotoxins (IT) are experimental pharmacologic agents that are made by linking antibodies or cytokines that specifically bind to cancer cells to potent catalytic toxins of which a single molecule can kill a cell. The major purpose is to deliver therapy selectively to cancer cells instead of nontarget organs as does conventional chemotherapy. Although these agents selectively bind and kill cancer cells, clinically they have been limited by their 1) failure to penetrate and localize in adequate concentrations in cancer target tissue 2) localization in nontarget organs limiting the tolerated dose and collapsing the therapeutic window. In this application, the applicant will explore a solution to this problem. Cells of the immune system such as T cells are the most prominent cell types that penetrate, attack, and destroy cancer cells and are naturally suited for the expression and production of cytokines in response to antigenic challenge. Therefore, he proposes using T cells to deliver retroviral IT (retIT) consisting of IL-4 spliced to genetically modified diphtheria toxin at the site of the leukemia cells. He has established a model of retIT therapy that he will use as a foundation for future attempts to modify and improve retIT. He will test the usefulness of retIT for therapy of myeloid leukemia, the most common adult form of leukemia. In this model, the applicant has produced an antigen specific CTL cell line called T15 by hyperimmunization with irradiated murine myeloid leukemia C1498. When T15 is transduced with retrovirus encoding IL-4 spliced to truncated DT, T15 cells have been shown to express and secrete IL-4 retIT which specifically kills IL-4R+ C1498 cells, but not IL-4R- cells in vitro. More importantly, mice given C1498 tumors show significantly enhanced anti-C1498 effects when treated with transduced T15 cells as compared to controls. In this application in the first aim, the applicant intends to use this model first to answer important questions regarding the role of IL-4 IT and T15 CTL in the first retroviral model. Is secretion of IL-4 IT necessary and how much must be secreted to get an anti-cancer effect? Does the amount of secreted retIT correlate with the magnitude of the anti-cancer effect? What is the role of the CTL vehicle in the retIT response? Can we enhance secretion and CTL delivery of retIT? In the second aim, he will determine if retIT administration has advantages over conventional IT administration particularly regarding their toxic effects on non-target organ systems and effects on the immune response. With the anti-C1498 effects that he has already established as a baseline, in the last aim he will ask whether or not important genetic modifications can improve retIT.